N2O/Ethanol bipropellant rocket engine

[stevep]

Hey Anders,

I took your word for it that you had a *high frequency* instability. I couldn't hear it on the video so I just assumed it was either my computer, my ears, or the video camera didn't capture it. Or, maybe we have different definitions of "high frequency" :-).
Anyway, given that you mention that the methanol was mixed with water, I've been doing some thinking and calculating (not always a good thing :-)).

Based on your comment that the methanol had 25% water in the first run, that would give you a C* of 1455 m/s. There is a simple formula to figure out chamber pressure given C*, throat area, and propellant mass flow: Pressure = (Cstar * mdot)/Athroat where mdot is in kg/sec, Athroat is the usual sq. meters, and pressure is Pascals. Since you don't have numbers for your propellant flow, I went back through the thread and found where you did the water flow tests.

The formula for mass flow through an orifice is mdot = Cd A sqrt(2 rho deltaP), and the usual problem is that one isn't certain of what the discharge coefficient (Cd) is. However, based on your water tests it looks like Cd = 0.7 is about right. Assuming Nitrous and watered-down methanol flow the same as water (probably not exactly true, but close enough for now), then at a pressure drop of 59 bar (60 bar -> atmosphere), you would have 0.42 kg/sec of nitrous through each of the 2 oxidizer orifices and 0.41 kg methanol/water through the fuel orifice for a total propellant mass flow of 1.25 kg/sec.

Inserting that number into the chamber pressure formula, I got a chamber pressure of 36 Bar which is higher than I think you were aiming for, but it would still give you almost a 100% pressure drop across the injector. However, now the chamber pressure is way above ambient (namely 36 Bar) so the pressure drop is only about 24 bar, which reduces the propellant flow accordingly. But even with only a 24 bar drop, you're still shoving 0.8 kg/sec of propellant through there which should give you a pressure of 23 bar and you still have more than adequate pressure drop across the injector.

But that assumes close to *perfect combustion* which almost no one ever gets on their first try. If your drop size and/or mixing were poor, you could have a very low chamber pressure. In solid motors (which is all I have first-hand experience with) chugging (*low*-frequency oscillation) is almost always a sign of poor combustion. If I just look at the exhaust, then I would agree with racket that you have a low-frequency problem because the mach diamonds shouldn't be moving around the way they are.

At this point, I'm not sure what to think. Given that we have only calculations to go by and no definitive chamber pressure measurements, it's hard to say what the next step should be. Given that things ran as well as they did (and I'm very impressed by the way), I might be tempted to just leave things as they are, run another test and make sure the camera on the gauges records the test. Then you would have chamber pressure to guide your next step. The other variable, of course, is the water in the methanol. Most methanol has a few percent water, but if you could get some with way less than 25%, and ran the next test with that, then there'd be one less potential issue to think about.

As it is, I'd hate to see you rip open your motor to fiddle with things that might not need to be changed. If it were me, I'd work really hard to get a test with chamber pressure and let that guide me from there.

As far as opening/closing the valves, some folks just use one strand of rope (to open) and have a strong spring close the valves.

The other things you have going for you (in terms of safety) is that the usual problem with liquid motors is burn-through rather than rapid disassembly. Hard startup I know you've thought about and it's the one I'd want to be well away from when it happened. Another thing to consider is no start (what about all that stuff that gets left in the chamber?). Probably just let the nitrogen purge everthing and then wait awhile, I guess. Nitrous + fuel *can* create explosive mixtures and with some water in the fuel, it might not all evaporate from the chamber (unknown to me, but I'd rather be safe). That's one argument for testing vertically--at least that way everything liquid drains out, but it does make the test stand harder to build/set up/hold down.

Oh yeah, one other thing: I don't know if your spark generator runs during the whole test (I know it gets ejected, but does it keep sparking?). Those things can cause problems with electronics, so your load cell data acquisition unit may get interfered with if the spark generator stays on (seems like *everyone* has problems with that).

Still, a very promising start and I have to say I'm envious!
--Steve

[Johansson]

Hi Steve, and thanks for the leghty reply!

The first run with 60 bar feed pressure I had filled the NOX tank with 2.3kg of nitrous oxide, judging from the video clip I would say that it ran for slightly longer than 10 seconds giving a NOX flow of only ~0.2kg/s.

This doesen´t add up at all with the calculated numbers, very annoying that I didn´t get a thrust chamber pressure reading since it would have helped a lot.

From the "looks" of the exhaust jet while running I would say that it seems to be in the ball park regarding propellant flow, the mach diamonds wouldn´t be there if something was way off the scale, right?

I have a supply of 100% pure methanol so I can easily try with unwatered fuel the next time, I only added water to it as an extra safety against chamber overheating. I´ve read that many early rocket engines used that 75/25 mix so I figured it would work for me as well.

With so much pressure drop over the injectors I could perhaps try to lower the feed pressure to 50 bar just to see how the engine reacts, increasing the pressure to 80 bar like we did for the second run only made things worse so perhaps less propellant mass flow would improve combustion? I totally agree with you that the wiser thing to do at this point is to run the engine unmodified a second time to collect more data before doing any serious modifications to it.

One last thing, the NOX tank is made from mild steel. The nitrous safety paper you told me about suggested that no oxidising metals should be used in contact with NOX since it could act as a catalyst, is mild steel tanks therefore a huge NO-NO or just a mildly unsafer choice than aluminum or stainless if you know what I mean?

In theory I should of course build a blast proof test cell with stainless high pressure tanks, fittings and pipes etc etc but I simply cannot afford it, so I will have to sort out the most important stuff and try to do this relatively safely on a shoe string budget.

Thanks for the encouraging words!

Cheers!
/Anders

[stoffe64]

hello Anders!
here is a link to a guy named michael fuchs, he has plenty of interesting projects including different rocket engines!
www.jetaerospace.org/Thruster/Deflagration/

[stevep]

Anders,

I am *very* pleased you weighed the nitrous! There is no such thing as too much data from a test! Two more bits of data would help: what was the approx outdoor temperature when you ran your test--I arbitrarily picked 5C corresponding to a nitrous density of about 880 kg/m3, but, as you know, the density varies widely with temp. Not enough to explain the difference between the calcs and your results, but we should at least try to get close. Also, can you tell me what the water capacity of your nitrous tank is? That is, if you filled it full of water, how many liters would it hold? I have software that purports to model the emptying process which can tell us what the internal temperature of the nitrous is and when it will turn to vapor in the tank.

You might want to look at the Aspire paper "The physics of Nitrous Oxide". There is some good info in there about tank filling.

Meanwhile, I'm going to go back through my calcs and see if I can spot any errors. I know one possible problem is that I may have misunderstood your water test--I wasn't sure I understood which numbers were measured and which were calculated. (It's on page 8 of this thread). What I thought you did was measure water flow at 10 bar through two 1.5mm orifices for 30 seconds and measured 3L. This would give 0.05L/sec for *one* orifice. Then you measured flow through the single 1.0mm orifice for 30 seconds and got 0.7L for a rate of .023L/sec. You then enlarged the 1.0 mm hole to 1.5 mm hole, thus giving you a 2:1 nitrous:fuel ratio, with a total water flow rate through the oxidizer holes of .1L/sec and a fuel flow rate of .05L/sec.

Do I have that right? If not, all the calcs will be off....

Meanwhile, I'll go back through my calcs and see if there's any obvious mistake. I did them both in English and SI units as a check, but I'm prone to error with this sort of thing so I may well have screwed something up. I'll post separately on safety....

--Steve

[stevep]

More on Nitrous safety...

First, I think of safety in rocketry as consisting of "layers" of safety nets--they work together such that if there's a hole in one net, you may be saved by a "backup" net.

The first net is knowledge. People are often killed/injured out of pure ignorance. Know thy failure modes! And, more importantly, know when you don't know what the failure modes are. Nitrous safety is interesting: on the one hand, there are thousands of people working with it every day in the medical, car performance, and hobby rocketry fields with a pretty good safety record. Yes, there are accidents, and the Scaled Composites one is very much worth our attention. But it is easy to come away from such accident reports thinking that Nitrous is as touchy and dangerous as pure nitroglycerin, which is clearly not the case.

A second aspect of knowledge that I just don't know how to convey in writing is how *powerful* some of these chemicals can be. I know from personal experience that when my first solid motor exploded, I was absolutely shocked that something so relatively small could produce so much force! And I'd seen lots of video of explosions, I'd heard stories, etc., but still, in person it was a very sobering experience. There was no damage (except to the rocket and test stand) or injuries, but without proper precautions, there would have been. For sure, if you think you'll be able to duck out of the way, you'll be wrong--these things can happen with *zero* warning and metal flies very fast. Just a note: if something blows on you, stay under cover for at least 30 seconds, and preferably a full minute. Material can take a long time to fall from a great distance.

I have read everything about Nitrous that I can get my hands on. I sure don't know everything and I'm always ready to learn more, but I think I at least have a handle on the subject. There are some good references at the end of the Aspire paper--especially the one by Rhodes. Many of these papers are online--if you can't find one, I can probably get you a link.

This brings me to the second layer of safety net: materials and procedures. If you look at the car racing use of nitrous, the medical uses, and the hobby rocketry hybrids, you can determine which materials seem to be safe to use (at least *as* they are used in those industries) and what procedures they follow to stay safe. Unfortunately, you can never have a *complete* understanding (leaving some holes in the net), but you can at least know what some safe materials/procedures are and try to use/follow those. Rhodes tested a lot of materials fairly rigorously and we know what the car guys use. For example, I was thinking that brass/copper (which some list as unsafe or at least unwise) for nitrous should be avoided--but then I noticed that a siphon tube used by the car guys was copper! And Rhodes tested copper without any problem. So what to think? And what are the tanks made of that are used in the medical field? Steel, I think. Also the tanks (and valves) supplied by gas suppliers -- what are they made of? (I don't actually know, but I plan to check before I get any nitrous in my system!)

As far as procedures go, things do need to be clean. Maybe not as clean as with LOX, but for sure no oil/grease anywhere. And I would be very careful about the o-ring material and valve seat material. Whether ball valves need a relief hole is an open question with me right now. One of the safety nets we have is "quenching distance"--apparently if metal is within 1/2" (12mm) of the nitrous, it will absorb enough heat quickly enough that the nitrous decomposition will be quenched. This is great news for folks like us with small feed lines because it provides safety that the guys like Scaled Composites with their much larger feed lines don't have. And we stay away from pumps. And do things remotely.

As far as your welded nitrous tank, I don't know what to say. Welding leaves "crud" and I don't know how that would interact with Nitrous--and remember, nitrous vapor reacts differently from nitrous liquid. I know that I'm personally going to be using an aluminum tank, but then I don't weld and I use aluminum "tanks" for solid rocketry so the fabrication is easy for me. And, of course, with steel, there's always the question of its exact composition--we don't always know exactly what material we're working with, so we don't really know how it compares with similar materials that have a good safety record. I guess the other factor here is that your current practice is to be standing next to your steel "run" tank when you're filling it from the scuba tank. That leaves a bit of a hole in the safety net (zero distance), on top of the hole of not knowing exactly how the steel/nitrous will interact. I'd be tempted to patch one of the holes (remote fill or switch to aluminum).

Which brings me to the third safety net: distance. All the amateur/hobby rocketry procedures incorporate this--launches happen by remote control. Igniters are installed at the last minute, away from spectators and non-essential personnel. Filling of hybrid tanks (which are flight-weight and therefore without the safety factor of transportable tanks) is done remotely. If ignition fails, dumping of the nitrous from the flight tank is done remotely or happens naturally over time (through the vent hole in the flight tank).

The great thing about distance is that it is cheap and easily obtained! As you point out, we can't all afford bunkers--but we can usually afford distance and so we knowingly trade "bunker" for "distance", patching one hole in the safety net with a different net.

Anyway, that's the way I look at it. I know you are conscious about safety--I noted your comment about keeping nitrous temperatures below room temp, and your concern with not having a burst disk in your scuba tank. That's good! And you seem very willing to learn--also very good. I think where I see the biggest hole in your safety nets is lack of experience with rocketry--and since everyone has to start from zero knowledge at some point, that just can't be avoided. Hopefully I can supply some of the necessary knowledge/awareness, and maybe there are others out there that can spot holes that I may miss.

--Steve

[jdw]

Love the www.jetaerospace.org/Thruster/Deflagration/ link. Tons of eye candy, especially the JA103. Very beautiful design. Props to
stoffe64 for the excellent link.

Salute,
Justin

[stevep]

Well, this is pretty embarrassing--apparently I am not to be trusted with anything as simple as a hand calculator :-). My only excuse is that when converting from bar to pascals, there's an awful lot of zeros involved :-)

I got tired of doing the calcs by hand, wrote a short program to do it and discovered that although I'd done the conversion properly when doing the water calcs, somehow I got in the habit of inserting an extra zero when doing the nitrous flow calcs. Normally that would show up as an "off by 10" error, but there's a square root involved so it wasn't obvious (to me).

Anyway, here's the new result:

Pressure drop of 60 bar, nitrous flow through one 1.5 mm orifice is (ta daa): 0.125 kg/second and for two orifices would be 0.25 kg/second which I consider within experimental error of the 0.2kg/sec flow seen during the test. Using the one orifice nitrous flow figure of 0.127, the total propellant flow would have been roughly 0.381 and would have resulted in a chamber pressure of about 11 bar.

Using a pressure drop of 50 bar, there would be a total propellant flow of 0.348 kg/sec (almost the same as at 60 bar) resulting in a chamber pressure of about 10 bar.

Again, those are theoretical numbers based on complete combustion.

I do agree with you, Anders, that things must be in the right ballpark or you wouldn't have as good a flame as you have. If actual combustion is close to what I calculate--even at 80% of what I calculate, you'd still have, I think, adequate pressure for relatively good running.

We really do need some actual chamber pressure numbers to get much further with things, I think. The one problem may be that by using a camera (24 fps?) to record pressure data from an analog gauge, oscillations may not be captured. Still, knowing what the average pressure is would be a great help.

--Steve ("always check my numbers")

[ernie wrenn]

A couple of factors:

Steel tanks rust and will develop pin holes and add wielding to the pot and the safety factor in nill. Can you use it... back to the net theory.

A wielded aluminum Tanks is NOT to be trusted! The transfer of 1 carbon atom can cause the tank to explode. Plus the safety factor of a wielded tank is "0" at best. There is no way to know the structural integrity of the wielded area vs the tank material.

All of the bottles we use are extruded for a solid puck and certified for 1800 psi working pressure/Approx 4500 rupture. The valves have a 3000psi vent safety.

Pickup tubes in the large tanks are copper tubing with a brass fitting in the valve. Remember nitrous must be treated like oxygen. No grease, oil or petroleum products should be enclosed in the nitrous area. PVC pipe WILL react and explode when pressurized in a nitrous bottle. There have been several incidents where this was done (stupid), it blew through the wall at the speed shop.

We can supply numerous bottle sizes from 10 oz to 20 lbs with or without valves along with safety vents, including pipe away's.

All of the solenoids we use today are stainless with Neoprene o'ring and seals. Brass valves were used until the early 90's.

All of our syphon tubes are aluminum.

Ernie
Compucar Nitrous Systems
1-800-NITROUS

[Johansson]

Hi Steve,

I think you were about right about the outside temp, 5°C one degree more or less.

I am not sure of the NOX tank volume, I know I have it written down somewhere but cannot figure out where. Does it really matter when I use compressed air to pressurise the tanks? The NOX shouldn´t turn to vapour as long as it is kept under a pressure well above its natural pressure.

thus giving you a 2:1 nitrous:fuel ratio, with a total water flow rate through the oxidizer holes of .1L/sec and a fuel flow rate of .05L/sec.


Exactly, at 10 bar pressure drop this is correct. Knowing from the test that the NOX tank flowed 2.3kg in 13 seconds at 5°C we have 0.18kg/s or 0.2L/s of nitrous flowing through the injector. What would the theroretical pressure drop over the injector be then when we know that 10 bar gives 0.1L/s?

About filling the run tank, would it be ok to raise the fill tank pressure with compressed air and then connect it (inverted of course) to the run tank to get the desired pressure difference between the two tanks? Much easier to do than heating/chiling the tanks or adding a vent to the run tank.

I´ll comment on the safety bit later, my daughter has passed a cold over to me so I need to get some sleep to be able to drag my sorry self to work tomorrow...

Cheers!
/Anders

[racket]

Hi Steve

I'm really enjoying your "numbers" :-)

One thing I'd also suggest is measuring actual thrust , its the best "cross reference" to check if an engine is performing as expected , its like have your car dyno'ed , if the horsepowers there then alls well :-)

As with our turbine engines, theres no way of knowing if the engine is doing what it should without the thrust being measured , we can then calculate backwards using the other measured running parameters to cross check that things are in balance .

LOL , I spent years trying to find thrust I imagined I should have been getting only to finally realise after doing the numbers that it was never going to be there ...........duh :-(

Cheers
John

[Johansson]

I was a bit slow there and didn´t see your last post Steve. The calcs looks more realistic now, a chamber pressure around 10 bar sounds much better since the water cooling doesen´t add any pressure on the outside of the aluminum chamber to counteract the pressure inside it like a regeneratively cooled chamber would. 30-40 bar chamber pressure with ambient pressure on the outside would probably burst the chamber...

Would my goal of 1000N thrust still be realistic or should I expect less? It didn´t "feel" like anywhere near that much even though one might get fooled by the non-flexing characteristics of the strain gauge compared to the spring gauges I am more used to.

Another test will be done as soon as possible, I´ll get back to you with a list of the things I think needs to be altered compared to the last run and hear what you think.

(And yes, the damn ignition was most likely the reason why the thrust reading got messed up. Next time I´ll separate the two as much as possible and have someone to shut off the power to it once the engine has lit.)

Cheers!
/Anders

[Johansson]

Hi Ernie,

Wielding aluminum tanks can be very dangerous, especially if you happen to hit someone in the head while waving it around. ;D

(Sorry about the joke, but you left an open goal there.)

I have pressure tested the tanks with water up to 120 bar so I consider them safe for the pressures I am using, sure it would be bullet proof to buy a certified tank but unfortunately bullet proof in the meaning that I cannot afford to continue working on the rocket engine...

Cheers!
/Anders

Nov 21, 2012 14:23:15 GMT -5 ernie wrenn said:

A couple of factors:

Steel tanks rust and will develop pin holes and add wielding to the pot and the safety factor in nill. Can you use it... back to the net theory.

A wielded aluminum Tanks is NOT to be trusted! The transfer of 1 carbon atom can cause the tank to explode. Plus the safety factor of a wielded tank is "0" at best. There is no way to know the structural integrity of the wielded area vs the tank material.

All of the bottles we use are extruded for a solid puck and certified for 1800 psi working pressure/Approx 4500 rupture. The valves have a 3000psi vent safety.

Pickup tubes in the large tanks are copper tubing with a brass fitting in the valve. Remember nitrous must be treated like oxygen. No grease, oil or petroleum products should be enclosed in the nitrous area. PVC pipe WILL react and explode when pressurized in a nitrous bottle. There have been several incidents where this was done (stupid), it blew through the wall at the speed shop.

We can supply numerous bottle sizes from 10 oz to 20 lbs with or without valves along with safety vents, including pipe away's.

All of the solenoids we use today are stainless with Neoprene o'ring and seals. Brass valves were used until the early 90's.

All of our syphon tubes are aluminum.

Ernie
Compucar Nitrous Systems
1-800-NITROUS

[stevep]

John: glad you like the numbers--more are on the way, see below... :-)

Ernie: Glad you chimed in with the nitrous info--been wanting to hear from someone in the industry. That's good info on the tanks, fittings, etc.

Regarding welded tanks, I notice that in the U.S., propane tanks (for home BBQs, forklifts, etc.) are made with welded materials, both aluminum and steel. The US Dept of Transportation (DOT) has standards that cover both. They require pressure testing of each tank during manufacture and inspections of samples and so on. Hydro testing of welded aluminum tanks is 2x service pressure; I didn't chase down the steel regulations.

I also note that my shop air compressor uses welded steel tanks.

Given that Anders tested his tank to 2x service pressure, I think he's in pretty good shape.

Instrumentation: it is true that the bottom line number of concern is thrust, so that definitely needs to be measured, and Anders has a load cell in the test stand to measure it. He just didn't get any numbers in this test.

However, rocket thrust comes from two components: chamber pressure which is then multiplied by the divergent portion of the nozzle. So if the thrust turns out to be low (as it usually does during testing), one is left with the question of where is the problem: pressure or the nozzle? While you can back-calculate things, these are always theoretical and do not always account properly for losses, so if you just have thrust, you're still (in practice) left wondering about the chamber pressure. Turns out that to stay sane, you want both pressure and thrust and you want some way of correlating the two (in time) because they're often varying during the test.

Also, having both means that you can use one to validate the other, at least to a ballpark level of accuracy. Finally, some instruments are more sensitive/responsive than others so one may reveal oscillations that the other doesn't record. Anders' load cell, for instance, presumably is sized for 1,000N of thrust--that's a pretty large number so it might not have the resolution to pick up an acoustic oscillation or the sampling period may not be short enough to do so. Also, his load cell is attached to the *cart* not the rocket chamber. That's a lot of mass, hence a lot of inertia, so I would expect to see only fairly gross resolution from it. That's fine for answering questions like "1000N or 800N?" but not so fine for less obvious problems like oscillation.

Often the test stand/instrumentation requires more time, energy, and money than the rocket motors. Sigh....

Motor casing strength: First, some mechanical engineering: a closed cylinder experiences stress in two directions: axially and around the circumference. The "around the circumference" is known as "hoop stress" (like the hoops that are used to hold the staves of a wooden barrel in place). It turns out that the hoop stress is always greater than the axial stress, so if the wall thickness is sized for hoop stress, it will be adequate for axial stress.

Hoop stress is a simple calculation for cylinders with "thin" walls (which the cylinders we usually deal with typically have--cannons have "thick" walls). The formula is

stress = (P * d) / (2 * t)

where P is pressure, d is the diameter of the tube, and t is the wall thickness, and stress is force/unit area. If we're doing the calc in English units, P is in psi, d is in inches, t is in inches, stress is in psi. Metric units are Pascals and meters. I'm going to do the calc in English units which are more convenient to deal with, then convert to metric at the end (blasphemy, I know :-)).

20 bar chamber pressure is 290 psi. Your chamber diameter (I think, based on a sketch earlier in this thread) is about 45 mm, or about 1.75" and you mention a wall thickness of about 3 mm (0.12"). So that gives us (for 20 bar) :
stress = (290 * 1.75) / (2 * .12) = 507.5 / .24 = 2115 psi

OK, what does this mean? It means that the aluminum will need to resist a force of about 2115 pounds / square inch to keep the cylinder from bursting. So how strong is aluminum? It varies, but the type known as 6061-T6 (a common high-strength type) is capable of withstanding 40,000 psi before it ruptures (this is called "ultimate strength") and about 35,000 psi before it even stretches (this is called "yield strength"). OK, back to metric in which it is common to report material strength in mega-Pascals (MPa). The stress of 2115 psi is equivalent to 14.6 MPa and the ultimate strength of aluminum is 276 MPa, yield strength about 241 MPa.

However, that is the strength of aluminum at room temperature! Aluminum loses strength quickly as the temperature rises: at 100C it still has about 93% of its ultimate strength but at 200C it has only about 30% and at 300C it is down to about 10% of its ultimate strength. Hence the need for cooling :-) But notice that your hoop stress is only 6% of aluminum's yield strength--you can therefore afford to let it get pretty hot before you have to worry about the chamber bursting.

The next question will be about how hot the aluminum will get if it runs for X seconds with a given quantity of water in the cooling jacket. Those are much more involved calculations, but they're entirely doable and I have software that does it. But I need accurate measurements of the water flow (liters/minute), exact thickness of the wall, and the thickness of gap between the chamber wall and the outer wall (and maybe a couple of other things, I don't remember them all offhand). However, based on intuition, I'd say you're pretty safe with your present cooling arrangement for short tests (less than 15 seconds).

Expected Thrust: This is pretty easy to calculate: F = P * At * Cf
Thrust (F, Newtons), P chamber pressure (Pascals), At (throat area, sq meters), Cf (thrust coefficient, dimensionless). That last item, Cf, requires explanation. Roughly speaking, if you had just a hole in the end of your thrust chamber, with no diverging section like your nozzle has, your thrust would be equal to P * At. But the diverging section amplifies the thrust by accelerating the exhaust gasses to more than the speed of sound. How *much* more depends on the nozzle's configuration and the composition of the gasses.

The primary factor affecting the nozzle's performance is, as I think you know from Sutton, the expansion factor: Area of exit / Area of throat. Your exit diameter is 45mm (from earlier in this thread) and your throat diameter is 25mm. Doing the math, this gives an *area* ratio of about 3.14. This is a bit shy of optimal which would be about 4.2. The relationship between area ratio and thrust coefficient is not linear, so you're more efficient than you might think from the above--I'll do some calcs and get back to you on that. But for now, let's say the Cf is about 1.2 whereas theoretical optimal would be about 1.4.

So lets use Cf = 1.2, P = 20 bar = 20,000,000 Pascals, Athroat = .000506 sq meters which gives us a thrust of 1177 N, at 10 bar you'd get 589 N. With a Cf of 1.35 and 20 bar you'd be up to 1325N, and at 10 bar it would be 662N. It all comes down to chamber pressure, and chamber pressure is all about mass flow of propellant.

At this point, I think the best thing is to rerun the test (assuming it is fairly easy to arrange), as you say you're inclined to do, and just see where things are at.

Oh, yeah, don't sweat the tank volume just yet--when I wrote that I forgot you were pressurizing with nitrogen. As this response is getting quite large, I will address the pressure drop over the injector in another post....

--Steve

[stevep]

And now for some not-so-heavy content, aka "advanced eye candy":

Many of you may have seen this stuff, but for those who haven't, here are some links videos/images/blogs from some teams that entered the "Lunar Lander Challenge" which was created back in 2006. Briefly, the point was to make a rocket that could take off vertically, hover, move laterally some distance, then land vertically, simulating what would be required of a real lunar lander. Prize money was substantial: for the highest-level achievement it was a million US dollars. More info on the challenge rules, etc. is here:

en.wikipedia.org/wiki/Lunar_Lander_Challenge

One of the entrants, Armadillo Aerospace started out in 2000 as a few guys in a garage working on their first liquid fuel motor. They did a blog for many years that was very detailed and instructive:

www.armadilloaerospace.com/n.x/Armadillo/Home/News

If you take the time to read through the blog, there are quite a few photos embedded in it.

They have some great pages with links to videos and images. You can also search YouTube for their videos. It is quite something to see a rocket lift off, hover, and land.

www.armadilloaerospace.com/n.x/Armadillo/Home/Gallery/Wallpapers
www.armadilloaerospace.com/n.x/Armadillo/Home/Gallery/Videos
www.armadilloaerospace.com/n.x/Armadillo/Home/Gallery/Images

Paul Breed (and his son) also decided to enter and have a very instructive blog (w/embedded photos) here:

unreasonablerocket.blogspot.com/

YouTube video of their first full vehicle test here:

www.youtube.com/watch?v=ajs9GKT96GA&feature=channel&list=UL

Many other videos available on his channel.

Masten Aerospace was the eventual winner, video here:

www.youtube.com/watch?v=qRFsGhti_D8&feature=related


Enjoy,
--Steve

[Johansson]

Hi Steve,

You don´t know how glad I am for your support in this, no matter how much I prefer the trial-and-error approach to things there is little sense in not backing it up with solid theory and some calcs to confirm the test results.

Regarding safety, just as you stated the main thing I lack is experience with these kind of engines. I am not overly concerned about nitrous tanks blowing up on me without an obvious reason but a "high" thrust bipropellant rocket engine like this inspires caution in a way that a gas turbine don´t, as long as you keep out of the plane of rotation you are fairly safe even if a turbine grenades on the test stand. Not so with a rocket engine...

Great to hear that the thrust chamber should cope with the pressure and temp, I haven´t got any numbers for the coolant flow but I trust your intuition on this one. The chamber looks as new and the cheramic coating is still there so I think it will be fine even if I skip watering the methanol.

I know where to borrow a digital pressure sensor that I can fit for the next test, then we can get the chamber pressure and load cell value on the same graph without having to video tape a pressure gauge. To avoid interference from the ignition coil I will make sure someone shuts the power off when the engine has started.

Not sure how soon I can have everything set for a new test though, it takes most of a day to do and I have heaps of boring things to take care of first like fix the car and paint a couple of rooms in the house...

1177N at 20bar sounds perfectly ok to me, I hope to see similar numbers during the next test!

Cheers!
/Anders